From: Joe Chang (changjoe@earthlink.net)
Date: Sun Jun 27 2004 - 20:08:49 GMT-3
Looking at my response I realize it was sort of rude, for which I apologize,
Peter.
If you think about all the IP routing protocols out there, you notice none
make path decisions based on QoS. I can think of one non-IP routing protocol
that does make its decisions based on QoS: ATM's PNNI.
I don't know much about MPLS, but I do know that MPLS replaces PNNI. Instead
of deciding by QoS, an MPLS-ATM network goes back to deciding by path cost.
So maybe multiple topologies via MPLS is not the solution.
I would say PBR is the simplest solution. If the source is the voice VLAN
then send it across one circuit, else use the other circuit.
----- Original Message -----
From: "Joe Chang" <changjoe@earthlink.net>
To: "Peter van Oene" <pvo@usermail.com>; <ccielab@groupstudy.com>
Sent: Sunday, June 27, 2004 6:03 PM
Subject: Re: policy routing via OSPF
> More than one topological view is not what I had in mind with PBR. But
it's
> a poor idea in the first place ...
>
> ----- Original Message -----
> From: "Peter van Oene" <pvo@usermail.com>
> To: <ccielab@groupstudy.com>
> Sent: Sunday, June 27, 2004 2:15 PM
> Subject: RE: policy routing via OSPF
>
>
> > how does pbr deliver more than one topological view? ie, one physical
set
> > of links, and two or more sets of best paths - which would allow per
> > source/dest best paths etc?
> >
> >
> > At 02:07 PM 6/27/2004, Nir Wittenberg wrote:
> > >Performance penalties and it is not dynamic.
> > >
> > >-----Original Message-----
> > >From: Joe Chang [mailto:changjoe@earthlink.net]
> > >Sent: Saturday, June 26, 2004 8:56 AM
> > >To: ccielab@groupstudy.com
> > >Subject: Re: policy routing via OSPF
> > >
> > >The OSPF RFCs detail an unimplemented TOS feature. I don't think OSPF
is
> > >the
> > >way to do it. On certain platforms ATM has a feature to split a flow
> > >into
> > >different VCs based on packet precedence, but I don't think that is
> > >available for FR.
> > >
> > >Your best bet may be PBR, but of course there are performance penalties
> > >with
> > >that approach.
> > >
> > >_______________________________________________________________________
> > >Please help support GroupStudy by purchasing your study materials from:
> > >http://shop.groupstudy.com
> > >
> > >Subscription information may be found at:
> > >http://www.groupstudy.com/list/CCIELab.html
> > >
> > >_______________________________________________________________________
> > >Please help support GroupStudy by purchasing your study materials from:
> > >http://shop.groupstudy.com
> > >
> > >Subscription information may be found at:
> > >http://www.groupstudy.com/list/CCIELab.html
> >
> > _______________________________________________________________________
> > Please help support GroupStudy by purchasing your study materials from:
> > http://shop.groupstudy.com
> >
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
> _______________________________________________________________________
> Please help support GroupStudy by purchasing your study materials from:
> http://shop.groupstudy.com
>
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Sat Jul 03 2004 - 19:40:51 GMT-3